domains / families / agent-tooling / global-agent-guardrails

global-agent-guardrails

separate

Configure the shared guard against catastrophic shell commands in local AI agents. Use when changing block patterns, adding an agent or machine, or investigating why a command was or wasn't blocked.

When to use

• changing block patterns
• adding an agent or machine
• investigating why a command was or wasn't blocked
• blocking catastrophic shell commands
• tuning shared patterns file
• configuring per-agent wiring
• running guard test suite
• E2E verification of guard

When not to use

• guarding against malicious agents
• blocking recoverable local commands
• over-blocking harmless commands

Skill page

Source